Summary: Overview The recent development of rapid molecular diagnostics promises to revolutionize TB control. Such tests can be performed directly on sputum or other clinical samples and rely on the rapid detection of Mycobacteria tuberculosis (MTB) genetic signatures as well as drug resistance- associated mutations. Clinical studies demonstrate that they perform better than sputum smear microscopy and reliably identify resistance to several first line drugs. However, these tools suffer from two important limitations. First, current platforms are limited in their ability to extract MTB DNA directly from clinical samples. And secondly, there are major gaps in our knowledge of the genetic determinants of poor treatment responses. Here, we propose to address these gaps through a multi- disciplinary collaboration emphasizing discovery of new biomarkers of drug resistance and tolerance, the identification of optimal clinical sampling strategies directed toward detection of MTB DNA and the development of a sensitive micro-array based rapid diagnostic. Our long-term goal is to develop a diagnostic strategy that will improve the diagnosis of DR TB and stem the further spread of the disease. The immediate aims of this projects are to 1) discover and characterize novel biomarkers of TB drug resistance both through human studies and by phenotyping isogenic strains with introduced mutations; 2) continue late stage development of a novel diagnostic with which to detect MTB DR, 3) develop the most promising clinical sampling strategies with which to detect DNA from MTB and 4) develop and optimize a micro-array based near Point of Care diagnostic designed to detect adult and pediatric MTB and its drug resistance mutations.